Stemness-Suppressive Effect of Bibenzyl from Dendrobium ellipsophyllum in Human Lung Cancer Stem-Like Cells

Cancer stem-like cells (CSCs) are key mediators driving tumor initiation, metastasis, therapeutic failure, and subsequent cancer relapse. Thus, targeting CSCs has recently emerged as a potential strategy to improve chemotherapy. In this study, the anticancer activity and stemness-regulating capacity of 4,5,4′-trihydroxy-3,3′-dimethoxybibenzyl (TDB), a bibenzyl extracted from Dendrobium ellipsophyllum, are revealed in CSCs of various human lung cancer cells. Culture with TDB (5–10 μM) strongly abolished tumor-initiating cells in lung cancer H460, H23, and A549 cells in both anchorage-dependent and anchorage-independent colony formation assays. Through the 3D single-spheroid formation model, attenuation of self-renewal capacity was observed in CSC-enriched populations treated with 1–10 μM TDB for 7 days. Flow cytometry analysis confirmed the attenuation of %cell overexpressing CD133, a CSC biomarker, in TDB-treated lung cancer spheroids. TDB at 5–10 μM remarkably suppressed regulatory signals of p-Akt/Akt, p-GSK3β/GSK3β, and β-catenin corresponding to the downregulated mRNA level of stemness transcription factors including Nanog, Oct4, and Sox2. Moreover, the antiapoptosis Bcl-2 and Mcl-1 proteins, which are downstream molecules of Akt signaling, were evidently decreased in CSC-enriched spheroids after culture with TDB (1–10 μM) for 24 h. Interestingly, the diminution of Akt expression by specific siAkt effectively reversed suppressive activity of TDB targeting on the CSC phenotype in human lung cancer cells. These findings provide promising evidence of the inhibitory effect of TDB against lung CSCs via suppression of Akt/GSK3β/β-catenin cascade and related proteins, which would facilitate the development of this bibenzyl natural compound as a novel CSC-targeted therapeutic approach for lung cancer treatment.


Introduction
Lung cancer has been recognized as one of the most common cancers worldwide, accounting for a large proportion of cancer deaths [1]. Comprising approximately 80% of all cases, non-small-cell lung cancer (NSCLC) is the predominant type, the majority of which are diagnosed at an advanced or metastatic stage [2]. Despite the progress in cancer research, there is still a low five-year survival rate among lung cancer patients [3,4]. Recurrence of tumor lesions resulting in high mortality rates indicates a failure of available therapeutic regimens for lung cancer [5,6]. Chemotherapy, especially when combined with other remedies such as surgery or radiation, has been shown to offer a survival advantage to NSCLC patients [7,8]. However, key barriers in chemotherapy such as innate or developed chemoresistance and high incidence of cancer relapse after completion of therapeutic regimens have been consistently reported [9,10].
It has been accepted that the heterogeneity of cancer population crucially influences the relatively low success rate of current anticancer drugs [11]. e subpopulation of cancer stem-like cells (CSCs) within tumor tissue is imbued with stemness features of self-renewal and multilineage differentiation, thus allowing tumor initiation from the remaining CSCs after chemotherapy [12]. CSCs that highly express CD133 (prominin-1) on cellular membrane isolated from both clinical specimens of lung cancer patients and human lung cancer cells have been documented to be mediators of drug resistance and tumor initiation properties [13,14]. Conventional chemotherapeutic agents, however, do not eradicate CSC subpopulations [15]. erefore, novel chemotherapeutic moieties targeting CSCs are urgently needed for improved lung cancer treatment [16]. e stemness phenotypes of CSCs are mediated by several pluripotent transcription factors including Oct4 (octamer-binding transcription factor 4), Sox2 ((sex determining region Y)-box 2), and Nanog [17,18]. Oct4 plays a critical role in the tumorigenesis, self-renewal, and pluripotency of CSCs obtained from different cancers [19,20]. Likewise, Sox2 is an essential mediator of self-renewal in embryonic stem cells (ESCs) [21]. e overexpression of Oct4 and Sox2 influences not only the differentiation of mesenchymal stem cells but also the tumor-initiating capacity of CSCs [22]. Additionally, Oct4 and Sox2 transcription factors responsible for self-renewal and pluripotency are found to mediate such activities through complex interactions with Nanog [23]. Accumulating evidence presents the reduction of Oct4, Sox2, and Nanog expression levels as an effective strategy to suppress CSC characteristics in lung cancer [24][25][26]. e modulation of CSC features and related proteins is also regulated by upstream signaling of PI3K (phosphoinositide 3-kinases)/Akt (protein kinase B)/β-catenin pathway [27,28]. Activation of Akt with consequent phosphorylation of GSK3β (glycogen synthase kinase 3β) results in the release of β-catenin from the GSK3β degradation complex [29]. e augmented level of β-catenin in turn stimulates the transcription of Oct4, Sox2, and Nanog [30]. us, the inhibition on Akt/GSK3β pathway efficiently decreases CD133 high -expressing cells and tumor-initiating activity in lung cancer [31]. Interestingly, various natural compounds have demonstrated anticancer activity, especially on CSCs via suppression of Akt-mediated stemness regulatory proteins [26,32].
Bibenzyl of 4,5,4′-trihydroxy-3,3′-dimethoxybibenzyl (TDB) extracted from Dendrobium ellipsophyllum has gained interest due to its induction of apoptosis and repression of epithelial to mesenchymal transition (EMT) through Akt inhibition in human lung cancer cells [33]. Since targeting Akt-related signal is a promising strategy for CSC-directed therapy in lung cancer [25], the previous findings on TDB have prompted further interest in the potential of this natural bibenzyl to target lung CSCs, a facet which has not yet been explored.
is study aimed to evaluate the suppressive effect of TDB and the relevant underlying mechanisms on CSCs in human lung cancer cells. e obtained information would support the further development of the bibenzyl from D. ellipsophyllum as a potential chemotherapeutic drug using a CSC-targeted approach for lung cancer therapy.

Cytotoxicity Assay.
Cell viability was examined by MTT colorimetric assay. Firstly, cells seeded at a density of 1 × 10 4 cells/well in a 96-well plate were cultured with TDB (0-300 μM) for 24 h. en, the cells were incubated with MTT (0.4 mg/mL) for 3 h at 37°C in a dark place. e supernatant was removed before adding 100 μL/well of DMSO to dissolve purple formazan crystal. e intensity of the formazan product was determined via spectrophotometry at 570 nm using a microplate reader (Anthros, Durham, NC, USA). e percentage (%) of cell viability was calculated from the absorbance ratio between treated and untreated control cells.

Nuclear Staining
Assay. Mode of cell death was evaluated through nuclear costaining with Hoechst33342 and PI. e cells were seeded at a density of 1 × 10 4 cells/well in a 96well plate for 12 h and then further treated with TDB (0-50 μM) for 24 h. After the incubation with Hoechst33342 (10 μg/mL)/PI (5 μg/mL) solution at 37°C for 30 min, apoptosis cells characterized by bright blue fluorescence of Hoechst33342 and PI-positive necrosis cells presenting red fluorescence were visualized and counted under a fluorescence microscope (Olympus IX51 with DP70, Olympus, Japan).

Spheroid Formation Assay.
e three-dimensional (3D) spheroid formation was carried out in culture media supplemented with 1% FBS in an ultralow attachment plate for the enrichment of CSC populations in cancer colony [34]. Briefly, cells were grown in a 6-well ultralow attachment plate at a density of 3 × 10 4 cells/well and 500 μL of 1% FBS supplemented media was added to maintain the cells every 3 days for 7 days. After repeating 2 cycles of the described anchorage-independent growth, the secondary spheroids were transferred into 24-well ultralow attachment plate as single colonies. CSC-enriched spheroids were treated with various concentrations of TDB and further incubated for 7 days. At days 0, 1, 3, 5, and 7, the spheroids were photographed using an inverted microscope (Nikon Ts2, Nikon, Japan). e spheroid size at the observation day was calculated relative to the size at day 0 and presented as relative spheroid size.

Detection of CD133 via Flow Cytometry.
e secondary CSC spheroids grown in media supplemented with 1% FBS with or without TDB were subjected to flow cytometry for evaluation of expression level of a CSC protein marker, CD133, on cell membrane. After 3 days of incubation, CSCenriched spheroids were harvested by centrifugation at 200 g for 4 min. e cells were washed twice with PBS (pH 7.4) and resuspended in PBS containing 10% FBS (incubation buffer) as single-cell suspension. Next, the cell suspension was further incubated on ice with an anti-CD133 antibody for 1 h. After removal of primary antibody and washing with the incubation buffer, Alexa Fluor 488-conjugated secondary antibody was added, and the cells were incubated on ice for another 30 min. e labeled cells were washed with the incubation buffer and further resuspended in PBS for measurement of CD133-associated fluorescence intensity via Guava easyCyte flow cytometer using InCyte 3.3 software (EMD Millipore, Billerica, MA, USA).

Limiting Dilution Assay.
In this study, a limiting dilution assay (LDA) was used to determine the effect of TDB on CSC populations possessing tumor-initiating capability [35]. Briefly, the cells were prepared in media supplemented with 1% FBS and seeded in a 96-well ultralow attachment plate at a density of 1, 10, 50, 100, and 200 cells/well. en, the cells were cultured with or without different concentrations of TDB for 14 days. e number of colonies formed was counted and photographed under an inverted microscope (Nikon Ts2, Nikon, Japan).

Clonogenic Assay.
e ability of cancer cells to proliferate and generate tumor colonies after treatment with TDB was also evaluated through clonogenic or colony formation assay [36]. Human lung cancer cells at a density of 1 × 10 5 cells/well in the 6-well plate were cultured with various concentrations of TDB for 24 h. After removing undetached cells by washing with PBS, the remaining cells were made into single-cell suspension. Next, the viable cells derived after TDB treatment were seeded at a density of 250 cells/ well in the 6-well plate and cultured for 7 days. e formed colonies were then counted after fixing with methanol and acetic acid (ratio 3 : 1) solution followed by staining with 0.05% w/v crystal violet in 4% formaldehyde.

Western Blotting.
e secondary CSC-enriched spheroids were treated with different concentrations of TDB for 24 h. After centrifugation, the treated spheroids were incubated with 1× radioimmunoprecipitation assay (RIPA) lysis buffer ( ermo Scientific, Rockford, IL, USA) supplemented with a freshly prepared protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN, USA) on ice for 45 min. e supernatant was collected for the determination of total protein content using bicinchoninic acid (BCA) protein assay kit (Pierce Biotechnology, Rockford, IL, USA). Equal amounts of protein were separated through 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). e proteins were then transferred onto 0.45 μm nitrocellulose membranes (Bio-Rad Laboratories, Hercules, CA, USA), which in turn was immersed in 5% nonfat dry milk in TBST (25 mmol/L Tris-HCl, pH 7.4, 125 mmol/L NaCl, 0.1% Tween 20) at room temperature for 45 min. e membranes were then incubated with specific primary antibodies at 4°C overnight. After washing with TBST for 5 min × 3 times, the membranes were further probed with HRP-conjugated secondary antibody for 2 h at 25°C. A chemiluminescent reaction with Immobilon Western chemiluminescent HRP substrate was used to detect the signals of specific proteins, which was quantified Evidence-Based Complementary and Alternative Medicine via analyst/PC densitometry software (Bio-Rad Laboratories, Hercules, CA, USA).

Reverse Transcription Quantitative Real-Time PCR (RT-qPCR).
Total RNA was extracted from CSC-enriched spheroids after treatment with different concentrations of TDB for 24 h. RNA samples were reverse transcribed into cDNA using RevertAid First Strand cDNA Synthesis Kit ( ermo Scientific, Rockford, IL, USA). cDNA was then quantified by measuring the absorbance at 260 nm using ermo Scientific NanoDrop One microvolume UV-Vis spectrophotometers ( ermo Scientific, Rockford, IL, USA). e expression levels of transcription factor genes (Nanog, Oct4, and Sox2) and housekeeping gene GAPDH in the spheroids were analyzed by reverse transcription quantitative real-time PCR (RT-qPCR) using the CFX 96 Real-Time PCR system (Bio-Rad Laboratories, Hercules, CA, USA). e RT-qPCR was performed using Luna Universal qPCR Master Mix (Bio-Rad Laboratories, Hercules, CA, USA). e designed primers used in this study are presented as follows: Nanog: forward: 5′-ACCAGTCCCAAAGGCAAACA-3′ reverse: 5′-TCTGCTGGAGGCTGAGGTAT-3′ Oct-4: forward: 5′-AAGCGATCAAGCAGCGACTA-3′ reverse: 5′-GAGACAGGGGGAAAGGCTTC-3′ Sox2: forward: 5′-ACATGAACGGCTGGAGCAA-3′ reverse: 5′-GTAGGACATGCTGTAGGTGGG-3′ GAPDH: forward: 5′-GACCACAGTCCATGCCA-TCA-3′ reverse: 5′-CCGTTCAGCTCAGGGATGAC-3′. e initial denaturation step was performed at 95°C for 3 min, followed by 40 cycles of denaturation at 95°C for 5 sec and primer annealing at 57°C for 30 sec. Analysis was performed using the comparative Cq value method. e relative expression of each gene was normalized against the housekeeping gene product. Briefly, 100 nM siRNAs in OptiMEM cell culture medium ( ermo Fisher Scientific, Waltham, MA, USA) were incubated with Lipofectamine 2000® mixture for 15 min at room temperature and then added to lung cancer H460 cells. After incubation at 37°C for 48 h, the transfected cells were further subjected to 3D spheroid formation, LDA, and clonogenic assay for evaluating the anticancer activity targeting CSCs of TDB. e silencing of Akt and p-Akt protein expression level was evaluated by Western blot analysis.

Statistical Analysis.
e data from three independent experiments are presented as means ± standard deviation (SD). Using SPSS statistical software version 22 (IBM Corp., Armonk, NY, USA), one-way analysis of variance and Tukey HSD post hoc test were performed, with statistical significance at p value < 0.05.

Cytotoxic Profile of TDB in Human Lung Cancer Cells.
MTT viability assay was used to evaluate the toxicity of TDB (Figure 1(a)) in human non-small lung cancer cells. After 24 h of treatment at 0-50 μM, TDB reduced viability in human lung cancer H460, H23, and A549 cells in a dosedependent manner (Figure 1(b)).
e relatively nontoxic effect of TDB was noted at 1-5 μM, whereas a significant reduction of %cell viability was observed at 10-50 μM compared with the untreated control cells. Because hair loss is one of the serious side effects induced by chemotherapeutic agents [37,38], the determination of half-maximum inhibitory concentration (IC 50 ) was additionally performed in both lung cancer cells and human dermal papilla cells (DPCs) to verify the safety profile of TDB. As indicated in Figure 1(c), the selective cytotoxicity of TDB against human non-small lung cancer cells was evidenced with IC 50 value at higher than 300 μM in DPCs, whereas lung cancer H460, H23, and A549 cells possessed IC 50 of TDB approximately at 136.05 ± 4.66, 164.29 ± 14.47, and 163.66 ± 15.14 μM, respectively.
To further clarify cytotoxic activity, TDB-treated lung cancer cells were costained with Hoechst33342/PI for detection of the mode of cell death. Apoptosis cells characterized by condensed chromatin and/or fragmented nuclei, which was indicated by bright blue fluorescence of Hoechst33342, were clearly observed in H460, H23, and A549 cells in response to 10 μM TDB treatment (Figures 1(d), 1(f), and 1(h), respectively). It should be noted that there were no necrosis cells stained with red fluorescence of PI in TDB-treated lung cancer cells. Corresponding to the cell viability results, the augmentation of %apoptosis was significantly noted in all lung cancer cells treated with TDB at a concentration of 10-50 μM (Figures 1(e), 1(g), and 1(i)).

Diminution of Tumor-Initiating Activity in TDB-Treated
Lung Cancer Cells. To evaluate the inhibitory effect of TDB on the frequency of tumor-initiating cells in cancer populations, the limiting dilution assay (LDA) was carried out in human lung cancer cells at various cell densities (1, 10, 50, 100, and 200 cells/well) [39]. e presence of tumor-initiating cells in human lung cancer H460, H23, and A549 cells was clearly indicated by the generation of cancer colonies under detachment condition of LDA at all cell densities (Figures 2(a), 2(c), and 2(e), respectively). Interestingly, the population of tumor-initiating cells in H460, H23, and A549    Evidence-Based Complementary and Alternative Medicine cells was significantly diminished in a dose-dependent manner after incubation with 1-10 μM TDB as presented in Figures 2(b), 2(d), and 2(f ), respectively. Strong suppression of tumor initiation was indicated by low colony number and absence of Hoechst33342-stained colonies in response to treatment of TDB at 10 μM, especially in lung cancer H460 and H23 cells. e clonogenic or colony formation assay is based on the capacity of a single cancer cell to regenerate a cancer colony. It is widely used for the determination of the effectiveness of cytotoxic compounds to eliminate proliferating tumor cells [40,41]. After culture for 7 days, crystal violet-stained cancer colonies initiated from single cells of H460, H23, and A549 were evaluated, and the representative images are sequentially presented in Figures 3(a), 3(c), and 3(e). Human lung cancer cells that were derived after culture with TDB (1-10 μM) for 24 h possessed lower capacity of colony formation compared with nontreated cells. e dramatic reduction of colony number was correspondingly observed in lung cancer H460, H23, and A549 cells treated with 5-10 μM TDB (Figures 3(b), 3(d), and 3(f ), respectively).
Taken together, these results reveal that TDB could restrain the tumor-initiating capacity of human lung cancer cells in both anchorage-independent and anchorage-dependent colony formation models. It is worth noting that the diminution on tumor-initiating cells was promptly observed at a nontoxic (5 μM) concentration of TDB.

Suppressive Effect of TDB on CSC-Enriched Spheroids of Human Lung Cancer Cells.
ree-dimensional (3D) spheroid formation or colonosphere assay is commonly used to evaluate the self-renewal feature of CSCs [42]. Herein, CSCtargeted anticancer activity of TDB was evaluated in secondary CSC spheroids of human lung cancer cells. Although Hoechst33342 nuclear staining showed no detectable cell death in CSC-enriched spheroids after culture with TDB (1-10 μM) for 7 days, TDB-treated spheroids of lung cancer H460, H23, and A549 cells were markedly smaller than untreated control counterparts (Figures 4(a), 4(c), and 4(e), respectively). Indeed, the reduced relative size of CSC spheroids was noted early on day 3 of the incubation with 5-10 μM TDB (Figures 4(b), 4(d), and 4(f )).

TDB Decreases CD133 high -Expressing Cells in Lung CSC-Enriched Spheroids.
Because CD133 is considered as a surface marker for lung CSC detection [43], the expression of CD133 was investigated via flow cytometry in H460, H23, and A549 secondary spheroids cultured with TDB for 3 days based on the significant alteration of relative size resulting from this treatment condition. Flow cytometry plots distinctly demonstrate that TDB (5-10 μM) could diminish the CD133 high subpopulations in lung cancer H460, H23, and A549 spheroids as separately presented in Figures 5(a), 5(c), and 5(e). CSCenriched spheroids of lung cancer H460, H23, and A549 cells comprised approximately 80% of CD133 high -expressing cells, whereas the major subpopulation in TDB-treated spheroids consisted of CD133 low -expressing lung cancer cells ( Figures 5(b), 5(d), and 5(f), respectively). e augmentation of CD133 low -expressing cells corresponds to the diminution of relative CSC spheroid size after TDB treatment (Figure 4). ese results suggest that TDB inhibits self-renewal activity in CSCs of various human lung cancer cells.

Stemness Transcription Factors Downregulated by TDB.
e transcription factors, Nanog, Oct4, and Sox2, are recognized as key mediators of CSC characteristics including pluripotency and self-renewal [44]. To investigate the modulatory effect on these transcription factors, mRNA levels of Nanog, Oct4, and Sox2 were detected by RT-qPCR in CSC-enriched H460 spheroids cultured with TDB (1-10 μM) for 24 h. As depicted in Figure 6(a), there was no difference in the relative mRNA levels of Nanog, Oct4, and Sox2 between the spheroids treated with 1 μM of TDB and untreated control. Intriguingly, the downregulation of these three transcription factors was remarkably detected in the treatment of 5-10 μM TDB. is finding indicates that the suppressive effect of TDB on stemness features in human lung cancer cells is associated with decreased levels of selfrenewal transcription factors including Nanog, Oct4, and Sox2.

TDB Modulates Akt/GSK3β/β-Catenin Signaling in Lung CSC-Enriched Spheroids.
To elucidate the underlying mechanisms of TDB related to cancer stemness, the expression of upstream regulatory proteins was investigated by Western blot analysis in CSC-enriched spheroids of human lung cancer cells. After culture with 5-10 μM of TDB for 24 h, there was marked diminution of expression levels of phosphorylated Akt (p-Akt), phosphorylated GSK3β (p-GSK3β), and β-catenin in CSC-enriched H460 spheroids (Figure 6(b)). Despite producing no significant alteration of downstream transcription factors, treatment with 1 μM TDB was observed to modulate Akt/GSK3β/β-catenin signaling in CSC secondary spheroids (Figure 6(c)).
As the previous study revealed that TDB modulates Bcl-2 family proteins in human lung cancer cells via mediating Akt signaling [45], the alteration of antiapoptosis Mcl-1 and Bcl-2 proteins was additionally examined in CSC-enriched spheroids treated with TDB. e significant reduction of Mcl-1 and Bcl-2 detected via Western blot analysis was observed in CSCenriched H460 spheroids incubated with 1-10 μM TDB for 24 h (Figures 6(d) and 6(e)). ese results strongly indicate the anticancer activity of TDB in targeting CSC populations of human lung cancer cells.

TDB Suppresses CSC Phenotype in Human Lung Cancer Cells Dependently on Akt Mediation.
To further verify the underlying machinery involved in the CSC-targeted effect of TDB on human lung cancer cells, Akt, an upstream regulatory molecule, was subjected to knockdown in lung cancer H460 cells using specific siRNA.
After 48 h of transfection, the significant downregulation of Akt and p-Akt protein levels was demonstrated via Western blot analysis in lung cancer H460 cells transfected with siAkt compared with both wild-type (WT) cells and simismatch control-(siCtrl-) transfected cells (Figure 7(a)). It is worth noting that there was no alteration of Akt and p-Akt expression level in siCtrl-transfected H460 cells compared with nontransfected wild-type cells (Figure 7(b)). Likewise, the incubation with 5 μM TDB obviously suppressed tumorinitiating activity (Figure 7(c)), colony formation ( Figure 7(e)), and enlargement of CSC-enriched spheroids in siCtrl-transfected H460 cells (Figure 7(g)). Surprisingly, the reduced Akt and p-Akt expression level efficiently restrained CSC-targeted anticancer activity of TDB as evidenced with no significant alteration of tumor-initiating cells (Figure 7(d)), number of formed colonies under attachment condition (Figure 7(f )), and relative CSC spheroid size in siAkt-transfected H460 cells culture with 5 μM TDB (Figure 7(h)). ese results suggest that TDB suppression of CSC phenotype in human lung cancer cells may partially be due to the Akt-dependent mechanism.

Discussion
CSCs, the rare subpopulations in tumor tissue, are characterized by stemness features of self-renewal and pluripotency, which contribute to tumorigenicity, chemotherapeutic resistance, and ultimately cancer relapse [46]. us, CSC-targeted therapy is of interest as a novel and promising strategy for cancer treatment evidenced by the success of CSC biomarkers-mediated drug delivery systems in both preclinical and clinical studies [47][48][49]. However, the key limitation of conventional chemotherapies is that these agents primarily target proliferative tumor bulk, but not tumor-initiating cells [46,48]. Natural products are widely recognized as sources of drug leads, especially in cancer therapy [50]. Garnering much research attention, natural compounds were recently reported to produce a suppressive effect on CSC phenotype and related regulatory pathways [24,32,51]. Due to the crucial role of CSCs in initiation and maintenance of lung cancer [52], the results herein revealing anticancer activity of TDB, a bibenzyl extracted from D. ellipsophyllum, against lung CSCs could guide the future investigations and innovations related to CSC-targeted therapy.
Tumorigenesis and self-renewal of various CSCs have been reported to be mediated by Akt/GSK3β/β-catenin pathway [53]. Activated Akt (p-Akt), a recognized upstream molecule triggering survival cascade, is likewise linked to the Wnt/β-catenin signaling, which also regulates CSC features. Phosphorylation at Ser 9 of GSK3β by p-Akt results in deactivation of GSK3β, which in turn stabilizes β-catenin to stimulate the expression of stemness-related genes [54]. Recently, the induction of Wnt/β-catenin signaling via PI3K/Akt/GSK3β cascade has been proposed to drive drug resistance in cancer cells [55]. In vivo evidence suggests that Akt-driven CSC enrichment is mediated by activation of the Wnt/β-catenin pathway through GSK3β phosphorylation [56]. It has been reported that tobacco smoke, an important risk factor for lung cancer, stimulates the Wnt/β-catenin pathway and enhances the CSC phenotype in lung cancer patients [52]. Interestingly, small molecules derived from natural products are deemed to possess much therapeutic potential and were even observed to inhibit CSC subpopulations and abolish tumor growth in vivo through inhibition of the Wnt/β-catenin pathway [57,58]. ese reports correlate with the observed downregulation of p-Akt along with the reduced levels of p-GSK3β (Ser 9) and β-catenin in TDB-treated CSC-enriched spheroids (Figures 6(b) and 8 Evidence-Based Complementary and Alternative Medicine 6(c)). Additionally, mTOR signaling, which is widely implicated in cancer pathology, mediates CSC formation and drug resistance through phosphorylated Akt at Ser 473 [59]. e revealed downregulation of p-Akt (Ser 473) in CSCenriched spheroids treated with TDB indicates some molecular insight into the lung CSC-targeted capacity of TDB. e effects of TDB on initiation of tumor growth and CSC self-renewal were revealed via clonogenic assay under both detached and attached conditions [35,36] and the 3D spheroid assay, the latter particularly approximating in vivo conditions and serving as a robust in vitro model for screening of anticancer compounds [60]. Under the detached condition, characteristics and cellular signals involved in the stem-like phenotype of 3D cancer spheroids considerably resemble in vivo solid tumor [42,60]. While animal studies will further validate antitumorigenic efficacy, strongly support the CSC-targeted potential of this bibenzyl compound. e suppressive activity on CSC phenotypes correlates with the diminished expression of Oct4, Sox2, and Nanog, the downstream transcription factors of β-catenin, in TDB-treated CSC spheroids (Figure 6(a)). Oct4, Sox2, and Nanog are recognized transcription factors mediating tumor transformation, tumorigenicity, and metastasis. ese transcription factors play crucial roles in sustained properties of self-renewal and pluripotency in embryonic stem cells [61]. e complex network governing coupregulation of Oct4, Sox2, and Nanog is also noted in many human CSCs derived from renal cell carcinoma [62], hepatocellular carcinoma [63], breast cancer [64], and lung cancer [26]. Recently, the suppression of CSC phenotype in lung cancer induced by alpha-lipoic acid was posited to be driven by the reduction of p-Akt level, which led to the depletion of p-GSK3β, β-catenin, and Oct4 [25]. Consistent with this finding, the diminution of self-renewal and tumor-initiating activity in lung CSCs cultured with TDB may also result from downregulation of Oct4, Sox2, and Nanog mediated through modulation of Akt/Gsk3β/β-catenin signal. Additionally, the loss of inhibitory effect of TDB on self-renewal and tumor initiation in Akt knockdown lung cancer cells (Figure 7) further verifies the Akt-dependent mechanism of TDB.
TDB, a bibenzyl extracted from D. ellipsophyllum, has been previously recognized for potent cytotoxicity [45], potential antimetastasis activity, and EMT suppression capacity in human lung cancer cells [65]. Furthermore, TDB presents a good safety and efficacy profile, which is evidenced by the lower IC 50 reported in various human lung cancer cells compared with normal cells (Figure 1(c)) and the alteration of apoptosis-regulating proteins leading to selective apoptosis induction in lung cancer cells treated with TDB [45]. It has been well-established that Akt also plays a pivotal role in apoptosis through the modulation of Bcl-2 family proteins [66,67]. e upregulation of Bcl-2 has been reported to promote cell survival and chemotherapeutic resistance in lung cancer cells isolated from patients. Moreover, the antiapoptosis Mcl-1 protein is also known to mediate the preservation of stemness features [68,69]. e initiation, growth, and progression of cancer lesions have been promoted by Akt-mediated Mcl-1 expression [59]. Cotargeting of Bcl-2 and Mcl-1 is thought to improve mitochondrial priming towards apoptosis, which could lead to  better therapeutic responses [70]. e decreased expression levels of Bcl-2 and Mcl-1 in TDB-treated CSC-enriched spheroids ( Figure 6(e)) and modulation of stemness-related pathways strongly suggest the promising anticancer activity of TDB particularly targeting lung CSCs.

Conclusion
In the present study, the novel anticancer activity of TDB, a bibenzyl from D. ellipsophyllum, towards CSCs of human lung cancer cells is revealed as evidenced by the suppression of tumor initiation and self-renewal properties associated with downregulation of Akt/GSK3β/β-catenin signal and related proteins (Figure 8). e obtained information would thus support the development of TDB as an effective CSCtargeted treatment for lung cancer.

Conflicts of Interest
e authors declare that they have no conflicts of interest.